Production site RAC redundant architecture deployment

We need the advise from Oracle Experts regarding our production site RAC redundant architecture deployment
For some business constraint, We have only 2-NIC Available for configure the redundant RAC Deployment
So are planed to bonding the available 2-NIC and after NIC bonding we got only one bonding interface of 1gbps speed.
We have configure the public and private ip address in that available single bonding interface.
network configuration
#bond0 (public network)
10.106.1.246 rac1
10.106.1.247 rac2
#bond0:1(vip network 10.106.1.251)
10.106.1.251 rac1-vip
10.106.1.252 rac2-vip
#bond0:11(private network 10.10.0.1)
10.10.0.1 rac1-priv
10.10.0.2 rac2-priv
#bond0:2(scan ip)
10.106.1.244 rac-scan
Oracle Setup Details
Oracle11g r2 RAC/GRID Software
RAC/GRID: 2-Node Database Cluster
Database Storage: ASM
Please find the review of our Hardware and Software
Hardware:
IBM Servers and SAN Storage
NIC Speed: 1Gbps
Software: Redhat Enterprise Linux5.5 64bit
Application Behavior:
1.) High insert/update/delete/select on single table.
2.) multiple session connect and disconnect.
Can you please conform that above architecture will support in production site?
what is the advantages and disadvantages with above architecture in production site?
Can you please suggest the right way of deployment the redundant RAC deployment in production site?

This is one case where I would say that you are playing a very dangerous game with your production system. You asked for Expert opinions and you have been informed that this is a VERY BAD IDEA!!!! While you think that it works, don't come asking about node evictions when your bonded nics get saturated. I can say that I am an expert having installed, configured and spent time troubleshooting more than 75 clusters (2-6 nodes) on some very impressive hardware. The "big one" was 250TB on a 3 node RAC on Sun 6900's (48 dual-core x 192GB main memory with 8 NICS using SUN IPMP and 8 HBA's for SAN connectivity ). When you start having "weird" issues, Oracle will not support your configuration. You will need to fix it before they even begin troubleshooting it. Tell your manager that unless they spring for the appropriate configuration they should execute the following command: "Alter manager update Resume;" because it is not "IF" it will fail, but "WHEN" will it fail. Trust me, you and your managers have put your system in a very precarious position.

Similar Messages

Oracle Timesten Geo-redundent architecure deployment queries

Hi,
We have 6 sites and each site connect with lease line.
There are 5 different physical Remote Site we want to deployed the active/standby timesten database on 5 sites
We want to create the READ ONLY LOCAL CACHE GROUP for each 5 Sites which is auto refresh from single central database site(2-Node RAC)
Data Modification always happen on Central Site.
Oracle timesten active/standby is in local LAN on each site but each site is connect through WAN to the Central Local site where the Oracle RAC Database resides
and configure the Cache Connect with each site for Auto refresh the Read Only Local Cache Group on each site.
ARCHITECTURE OVERVIEW
Remote Site Location-1 ==>(Read Only Local cache Group )Oracle TT A/S<<====== Auto Refresh <---------||
||
Remote Site Location-2 ==>(Read Only Local cache Group )Oracle TT A/S<<====== Auto Refresh <---------|| (CENTRAL LOCATION SITE)
||
Remote Site Location-3 ==>(Read Only Local cache Group )Oracle TT A/S<<====== Auto Refresh <---------|| (2-Node RAC Database)
||
Remote Site Location-4 ==>(Read Only Local cache Group )Oracle TT A/S<<====== Auto Refresh <---------||
||
Remote Site Location-5 ==>(Read Only Local cache Group )Oracle TT A/S<<====== Auto Refresh <---------||
We have go through below Oracle Technical Paper.
http://www.oracle.com/technetwork/database/performance/wp-imdb-cache-130299.pdf
please check the Figure 7==>Incremental Autorefresh of Read-Only Local Cache Groups
Please suggest your advice and suggestion for above architecture deployment
Regards
Hitgon
Edited by: hitgon on Jul 10, 2012 6:44 PM
Edited by: hitgon on Jul 10, 2012 6:46 PM

There is a great deal of stuff you need to understand and bear in mind when implementing any complex system and this is n oexception. You should start by studying the TimesTen documentation. In particular you should familiarise yourself with the informaiton in the Cache User's Guide and the replication Guide. These contain lots of very important information regarding the setup/deployment and operation of A/S pairs and cache groups, configuring cache connect for RAC etc. You should also read through the troubleshooting guide to familiarise yourself with what things to look at if things do not seem to work as expected.
If you are not already very familiar with TimesTen I would also strongly recommed that you take the time to read the rest of the documentation. TimesTen is not Oracle Db and while it is very compatible with Oracle in many areas there are also a lot of significant differences which you need to take into account when developing applications, managing the system etc. if you are able to take an OU training course on Timesten then I would recommend that but if not then reading the documentation is a good second best.
We do not 'recommend' operating systems specifically but 64-bit Linux is certainly a good choice. You might like to consider Oracle Enterprise Linux instead of Redhat; it has some advantages.
Within each site both nodes in the TimesTen active/standby pair should be on the same LAN. I would recommend GigaBit ethernet as a minimum.
While it is possible to write your own scripts to handle deployment, monitoring, failover and recovery of A/S pairs it is much, much easier (and much more robust) if you deploy Oracle Clusterware at each site to provide fully automated management of the A/S pairs. That is our very strong recommendation and is also best practice.
From a TimesTen perspective, system clock synchronisation is only needed within each site (i.e. the system clocks on both nodes in an A?S pair need to be closely aligned). However, it may be desirable to have all the nodes in all the sites have their clocks aligned for other reasons.
You need to ensure that the bandwidth and latency of the WAN connections is adequate for the amount of refresh traffic that you will have. There is no easy way to estimate/calculate this; you will need to determine this empirically.
Those are probably the most important things. As you progress then you can of course ask questions in this forum and use Oracle Support.
Chris

ACE at production site, CSS at DR

Hi
We have ACE modules at our primary production site in an active/standby config - and have been working fine for some time.
eg.
context ace1
description context1
allocate-interface vlan 10
allocate-interface vlan 11
context ace2
description context2
allocate-interface vlan 20
allocate-interface vlan 21
There is a bridge group in each context and the ip range for context1 is 10.1.1.x and context2 is 10.2.2.x.
Rservers and VIPs in context1 are therefore in the 10.1.1.x range; in context2 they are in the 10.2.2.x range.
For a DR site, we want to use a CSS11506 to act as the load balancer in case of a disaster at the production site.
Is it possible for the CSS11506 to be configured to act the same as the ACE in case of a DR scenario? ie. Can a CSS be configured to so the same job as the ACE, maintaining the IP addresses assigned to the rservers/vips/vlans?
We need to have the IP addressing at DR to be the same the production site (vlans can be created at DR as required and are independent of the production site) - however the DR site is only 'active' when the production site is destroyed (or in testing obviously).
What would a sample config on the CSS11506 look like given the above ACE configuration?
(It's my first attempt at configuring a CSS!)
Cheers guys
Cameron

hi,
It is possible to have a css to do the same functionality. The context principle does not exist but you have the similar interface/vip redundancy see:
http://www.cisco.com/en/US/docs/app_ntwk_services/data_center_app_services/css11500series/v8.20_v8.10/configuration/redundancy/guide/redundgd.html
Bridging can also be done see:
http://www.cisco.com/en/US/docs/app_ntwk_services/data_center_app_services/css11500series/v8.20/configuration/routing/guide/rtbrgdgd.html
Keep in mind that the policies are replaced by owner & content rules, serverfarms do not exist, real servers are called services and predictors are callled loadbalancing systems:
http://www.cisco.com/en/US/docs/app_ntwk_services/data_center_app_services/css11500series/v8.20/configuration/content_lb/guide/cntlbgd.html

Can we move InfoPath forms from development site to production site?

From the question I mean only the form, not the site itself.
Is there any way to move a InfoPath form from SharePoint development site to production site?
Thanks,

just publish it to production site
http://office.microsoft.com/en-in/infopath-help/move-a-published-form-template-to-a-new-sharepoint-library-HP010102446.aspx
Step 1: Publish the form template to the new document library
Open the working copy of the form template in design mode.
On the File menu, click Publish.
In the Publishing Wizard, click To a SharePoint server with or without InfoPath Forms Services, and then click
Next.
In the Enter the location of your SharePoint or InfoPath Forms Services site box, type the location of the SharePoint site that contains the new document library that you want to move your form template to, and then click
Next.
On the next page of the wizard, click Document Library, and then click
Next.
Click Create a new document library, and then click
Next.
On the next page of the wizard, type a name and description for the new document library, and then click
Next.
Choose the fields in the form template that you want to appear as columns in the default view of the document library.
How?
Click Next.
On the next page of the wizard, verify that the information shown is correct, and then click
Publish.
To open the document library and test the process for filling out a form that is based on this form template, select the
Open this document library check box, and then click Close.
The document library will open in a browser. When you click New on the list toolbar, a form based on this form template will open in Microsoft Office InfoPath.
http://sharepoint.stackexchange.com/questions/17383/infopath-change-publish-location
If this helped you resolve your issue, please mark it Answered

Softwares/licenses require over Office 365 Development,QA, UAT and Production site

Hello, Client has Office 365 E3 version, this plan provide Development site. We require to do extreme customization and
third party integration for that we are planning to go with Provider hosted app.
So, what are the additional software’s/licenses would require for Development,
QA, UAT and Production site for 6 developer?
How can we manage Development licenses cost of Visual Studio 2013, SharePoint 2013 Designer for 6 developers?
Do client need to purchase additional licenses of Office 365 E3 for QA and UAT site?
Do client need to buy additional license for Azure Web Site or is it provided in Office 365 E3 version by default, as we need this for Provider hosted app?

Hi,
Per my understanding, there might be some queries on the development environment set up in Office 365.
About the requirement of setting up a development environment for Office 365, steps in details can be found in the two links below:
Setting up your Development Environment for O365 & Azure Development
http://blogs.msdn.com/b/steve_fox/archive/2013/02/03/setting-up-your-development-environment-for-o365-amp-azure-development.aspx
How to: Set up an environment for developing apps for SharePoint on Office 365
https://msdn.microsoft.com/EN-US/library/office/fp161179.aspx
For the license related issues of Office 365, I suggest you post it to Office 365 forum, you will get more help and confirmed answers there.
http://community.office365.com/en-us/f/default.aspx
Thanks
Patrick Liang
TechNet Community Support
Please remember to mark the replies as answers if they help, and unmark the answers if they provide no help. If you have feedback for TechNet Support, contact
[email protected]

RAC + dataguard architecture/setup metalink notes

Looking for RAC & dataguard configuration for one db ...

Hi,
I think it is rather worth to look around on the [Oracle Maximum Availability Architecture site|http://www.oracle.com/technology/deploy/availability/htdocs/maa.htm] for whitepapers than checking the Metalink (Well, at least I like MAA much better the Metalink)
Hope this helps.
Regards,
Jozsef

RAC defined architecture and NFS

Hi Friends, how are you all?
Carrying on with my fight with Grid Infrastructure package installation, after to research the internet I found a whitepaper signed by Oracle that treats about the use of Grid 11.2 files over NFS partitions/disk (I'll let the link to this doc below). The environment's architecture I'm working with, which was defined by a consultant, have two VMs and one of them sharing a disk (the entire /dev/sdb) with the another machine through NFS; and then, we have a global shared disk area as a storage to OCR, VF e database files into the RAC. In the midst of installation, I'm using Normal Redundancy where I state the three /ocr0x directories to OCR files and the three /vtf0x to Voting Files, as I explain in more details below.
+Oracle Clusterware 11g Release 2 (11.2) – Using standard NFS to support a third voting file for extended cluster configurations+
+=> http://www.oracle.com/technetwork/database/clusterware/overview/grid-infra-thirdvoteonnfs-131158.pdf+
The exported (NFS) disk partitions are:
/u01 *(rw,sync,no_wdelay,insecure_locks,no_root_squash) # -> $ORACLE_BASE, $ORACLE_HOME
/u02 *(rw,sync,no_wdelay,insecure_locks,no_root_squash) # -> $GRID_HOME
# ocr files
/ocr01 *(rw,sync,no_wdelay,insecure_locks,no_root_squash) # -> ocr files /ocr01/disk01, disk01 is the file OUI will delete
/ocr02 *(rw,sync,no_wdelay,insecure_locks,no_root_squash) # -> ocr files /ocr02/disk02, disk02 is the file OUI will delete
/ocr03 *(rw,sync,no_wdelay,insecure_locks,no_root_squash) # -> ocr files /ocr03/disk03, disk03 is the file OUI will delete
# ocr files
/vtf01 *(rw,sync,no_wdelay,insecure_locks,no_root_squash) # -> vtf files /vtf01/disk01, disk01 is the file OUI will delete
/vtf02 *(rw,sync,no_wdelay,insecure_locks,no_root_squash) # -> vtf files /vtf02/disk02, disk02 is the file OUI will delete
/vtf03 *(rw,sync,no_wdelay,insecure_locks,no_root_squash) # -> vtf files /vtf03/disk03, disk03 is the file OUI will delete
** On that whitepaper I found just to investigate if the manner I am using to share disks with NFS is correct, you may find some relevant options you must consider when mount disks whenever you use Grid + NFS.
Well, after many efforts and expend many times to have this RAC up & running, I started to receive the same errors when OHASD tries to start up. The error is "+...Inappropriate ioctl for device...+" - I started to "distrust" or "wonder" that something was so strange with some points on the defined architecture. Even running the runcluvfy.sh and having it finish with any one "failed" item, I've gotten problems with the installation since I started with it.
So, the paper I found, signed by Oracle said:
+The first Oracle Clusterware version to support a third voting file mounted using the standard NFS protocol is Oracle Clusterware 10.2.0.2. Support has been enhanced to Oracle Clusterware 11.1.0.6 based on successful tests. Using standard NFS to host the third voting file will remain unsupported for versions prior to Oracle Clusterware 10.2.0.2. All other database files are unsupported on standard NFS. In addition, it is assumed that the number of voting files in the cluster is 3 or more. Support for standard NFS is limited to a single voting file amongst these three or more configured voting files only. This paper will focus on Grid Infrastructure 11.2.0.1.0, since starting with Oracle Clusterware 11.2.0.1.0 Oracle Automatic Storage Management (ASM) can be used to host the voting files.+
=> When that stated document said "+All other database files are unsupported on standard NFS+", what does it mean? OCR is one of the unsupported files to be located upon NFS disk/partition?
I'm wondering that the defined architecture won't support what we're doing. Do you guys know about any case similar, where a two nodes RAC was implemented over a NFS disk? Do you know about any case that OCR, VF e database files were stored upon NFS partitions/disks? Is that mentioned architecture I'm using valid? Could you give some hints?
I used the forum to discuss some questions around this Grid implementation:
=> Re: G.I. install keeps failing @ root.sh ( specifically when Starting ohasd )
=> Re: Problems in downloading Oracle Linux [4 - 6]
Thank you in advance for any hint you all can give me this time! I'll appreciate any kind of help.

Hi,
I didn't fully understand your environment. You have 2 virtual nodes, right? Let's say node1 and node2. Now, where is your NFS storage and what do you plan it will contain?
Regarding Oracle documentation, there are 2 NFS "vendors", one is standard NFS (exported by any linux/unix machine), the other one is vendor supported NFS (such as Netapp, EMC, etc.).
Oracle RAC files (OCR, voting and database) are supported on the second one only. It is not supported to use standard NFS for database files. However, if you wish, you can put 1 voting on a standard NFS (this is for cases where you have 2 locations for the RAC and you want to have another voting in a different site).
In any case, if I understand correctly you have node1 exporting NFS to node2 for the RAC. If this is true it is not supported, and generally a bad idea because you lose your entire high availability (if node1 fails, you got nothing).
HTH
Liron

Active Directory Sites and Exchange 2013 Deployment

I've recently took over responsibility of an Exchange 2013 Organization that is deployed as follows:
Active Directory consists of 4 Sites. AD Site A, B, C, D Exchange 2013 Enterprise resides in 2 of the 4 AD Sites as follows:
AD Site A - ExchangeServer 1 and ExchangeServer 2
AD Site B - Exchange Server 3
AD Site C - No Exchange Servers
AD Site D - No Exchange Servers
All 4 AD Sites are 4 different Physical locations/datacenters. All 3 Exchange 2013 servers are multi-role servers.
The Forest in which Exchange resides in consists of an empty Root domain, a Production (child) domain and a Test (child) domain. Exchange resides in the Production (child) domain.
Issue: AD Site A contains DC's from all 3 domains: Root Domain, Production child Domain (this is where Exchange lives) and Test child Domain. I notice that Exchange in AD Site A is using DC's from the Root Domain for it's "DefaultGlobalCatalog",
"DefaultConfigurationDomainController" and "DefaultPreferredDomainControllers" This to me does not seem to be very efficient as any Address Book queries will have to be referred to by the Root Domain DC's to the Production child domain
where Exchange lives. All of the AD User accounts and mailboxes are in the Production child domain.
In a situation such as this, would it be advisable to build 2 additional AD sites specifically for Exchange? Rather than re-IP Exchange or risk the impact of moving several other (non exchange) servers to another AD site, I would add the IP address
of the Exchange servers /32 to the new Exchange dedicated AD Sites and erect a DC in these new sites adding its IP address /32. Any thoughts on this idea? If the subnet that exchange resides on is (for example) 10.60.3.0 /16 in AD Site A, and
I build a new AD site for Exchange and add the IP address of the Exchange server such as 10.60.3.141/32 for this new Exchange AD Site boundary, I can still leave the 10.60.3.0 /16 unaffected in AD Site A, correct?
I'm looking for Microsoft's best practices in terms of laying out AD and domain controllers pertaining to Exchange server 2013.

Hi Anthouyray,
Thank you for your question.
We could use the following command to exclude domain controller which is root domain controller:
Set-ExchangeServer –Identity <exchange servername> -StaticExcludeDomainControllers <root domain controller>
Then we could restart the service of “Microsoft Exchange Active Directory Topology” to check if the issue persist.
If there are any questions regarding this issue, please be free to let me know.
Best Regard,
Jim
Please remember to mark the replies as answers if they help, and unmark the answers if they provide no help. If you have feedback for TechNet Support, contact [email protected]
Jim Xu
TechNet Community Support

Data Guard two Production sites

I am exploring the best (and least expensive) way to configure Data Guard in our environment.
Has anyone had two production environments, each in a different location and set up Standby Data Guard databases across each.
For example -
Site 1 - Prod RAC 1 with database 1 and a standby for database 2
Site 2 - Prod RAC 2 with database 2 and a standby for database 1
Is there any load to the resources in either cluster when also maintaining a standby database along with production?
I searched Metalink and was not able to find information on this type of configuration.
Thanks!

Yes, your understanding is correct. I simplified our environment for the question, actually we have two sites in different geographic locations.
Site 1 is production and Site 2 is Development, but we want to protect each site for Disaster Recovery, etc.
Site 1 has a three node rac cluster with a 'PRD_A' database and Site 2 has a three node rac cluster with a 'PRD_B'.
So, I would like to create, just as you said, a standby for PRD_A in the Site 2 cluster & create standby for PRD_B in the Site 1 cluster.
And, actually, each site has 2-3 databases running that I would like to create standby's for. This is why I am wondering if anyone has done this and if the load would be an issue. Yo Yo got it ;-)
See PRD_A has 3 instances let suppose PRD_A1, PRD_A2, PRD_A3.
PRD_B has 3 instances let suppose PRD_B1, PRD_B2, PRD_B3
So for PRD_A you can create either one standby database as PRD_A_STBY.
or you can create three standby databases as PRD_A1_STBY, PRD_A2_STBY, PRD_A3_STBY on CLUSTER B
For PRD_B you can create either one standby database as PRD_B_STBY.
or you can create three standby databases as PRD_B1_STBY, PRD_B2_STBY, PRD_B3_STBY on CLUSTER A
But Recovery process will be performed any one of the standby of one database, Remaining standby databases will sit idle.
(or)
You can have only one standby database, So the database itself clustered so you can add instances if in case of Switchover/Failover.
Here you will have two database on each instance.
Here the things
1) how much RAM
2) is it server capable to hold two instances?

IP source gurad on redundancy architecture

Architecture: 2x C3560 L3 , 1x C2950-48
Symptom: 1.C2950 has a link to each C3560 for redundancy .
2.IP DHCP snooping and source gurad are enables on C3560.
Question: When active C3560 is failed, all traffic will be transfere by HSRP, but on standby C3560, it doesn't have ip dhcp snooping binding table for dhcp souce guard reference, it will cause all traffic to be dropped by PACL. For this, do you have any idea to solve this failover problem ?
Thanks

Hi,
If you have a very specific need in terms of natting and using a source group, you may want to consider applying a nat or source group via an ACL.
So if you only have 2 servers on the back end that you need to nat going outbound, you could configure an acl and apply it to the vlan it is inbound on and apply the source group on the acl. Here is a link on using acls:
http://www.cisco.com/univercd/cc/td/doc/product/webscale/css/css_750/cmdrefgd/cmdaclc.htm
acl 1
clause 10 permit 10.1.1.1 255.255.255.255 destination any sourcegroup natserver
assuming the 10.1.1.1 is the server you want to nat and natserver is the name of the source group.
Just beware of using acls as they can be tricky and please understand that there is an implicit deny on all vlans when you enable the acls, so make sure you have atleast a "permit any any destination any" with all vlans applied to it..
Regards
Pete..

Reroute some vrf traffic between 2 sites over redundant link

hey guys,
We have a single client (in vrf) with 2 sites in different states and running over our mpls core.. Our primary link in our core is experiencing degredation of service and want to route this client over our redundant link while keeping all other clients going over our primary link - is this possible?
The client in question has its own vrf (L3VPN) at both sites and is running over mpls between both sites. We want to re-route this particular client to take our backup path, while keeping everyone else between both sites going over the primary. We are not using TE, instead LDP to build MPLS.
I don't believe this is possible to only re-route one client, however I thought I would ask the question.
We cannot failover to secondary link for everyone between both sites because the link doesn't have the capacity.
Thanks in advance.

Hi,
Using MPLS TE would certainly be an option. You would need to setup an MPLS TE LSP over the backup. You would also need to configure a separate lookback interface on each PE and use this loopback interface address as the next hop for the specific VRF
ip vrf X
bgp next-hop loopback 999
ip route 255.255.255.255 Tu1
This way you would make sure that only the traffic for this specific VRF would travel over the TE tunnel.
Regards

How to get sideload product key in order to deploy windows store app into Windows 8 pro and windows RT

From where can i get the Windows 8 sideload product key and the process how to get it .please explain in detail .

I was able to resolve my issue once I worked with Microsoft Professional Services. Here are the steps needed to deploy a Windows 8 Store app for anyone else looking for help on this.
Create your app package in Visual Studio
    https://technet.microsoft.com/en-us/windows/jj874388.aspx
Create an app certificate using PowerShell as administrator, here is one way to do it.
    https://msdn.microsoft.com/en-us/library/windows/desktop/jj835832%28v=vs.85%29.aspx
Install the app on the device
    https://msdn.microsoft.com/en-us/library/windows/desktop/jj835832%28v=vs.85%29.aspx
Ensure Device has the correct GP settings
    http://blogs.technet.com/b/scd-odtsp/archive/2012/09/27/windows-8-sideloading-requirements.aspx
Install the application on the device
    Refer to Sideload the app for user section
    https://technet.microsoft.com/en-us/windows/jj874388.aspx
* If you run your app now you should get an error, this is because you have not installed a Sideloading Key
And the missing piece of the puzzle What is a Sideload key and how can I get one?
* Documentation on this is confusing so here is what I needed to know and found out
    Sideload key is a product key that is needed to be added to the device so that it can use Windows 8 apps that are not from the Store
    It is not a requirement to have the machine joined to the domain or have Windows Enterprise installed on it. In my case my target was a Windows 8.0 Pro tablet.
    You do not need Windows 8.1 to use sideloading it will work on Windows 8.0 as well
    You can purchase a key from a Third Party Distributor but you will need to know what to ask for
    Ask for a Windows 8 Sideloading Key, Sometimes referred to as a Windows 8 Embedded Sideloading product key
    They cost about $100.00 a pop but they sell them as a Volume License so the minimum you can by is 5.
    However in my case it turned out that my company had an Enterprise Sideloading License, I just needed to know who to ask and what to ask for.
    Once you get your xxxxx-xxxxx- Slideloading key from the distributor here are the steps to installing it on the device (Section 2 Step 3)
    http://blogs.msdn.com/b/micham/archive/2014/05/30/sideloading-store-apps-to-windows-8-1-devices.aspx
Once I installed that my application worked.

XI scalability architecture - deploying integration builder objects

Hi
I read in the SAP document - 'Scaling up XI' that the way to scale up XI is simply addition of dialog instances and all these additional dialog instances would access the database in the central instance.
Now, in this scaled up architecture, do we have to transport integration builder objects onto each of the additional dialog instances - or do the entire set of integration builder objects needed at runtime - reside on the database on the central instance and are accessed by the different dialog instances ? ( In this case, the integration builder objects have to be transported to the central instance only . and the CMS setup should point to the url of the central instance only )
Is there any document that clarifies this concept of moving the production objects onto the central instance ?
Any thoughts, ideas shared on this would be rewarded.

Hi Karthik,
U must ve read this document
https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/c3d9d710-0d01-0010-7486-9a51ab92b927
Scaling of XI servers is done when "Performance" is the main issue to be considered. scaling thus generally includes the scaling of IS. So the entire set of IB objects should be placed at central instances only
This will help u
/people/sriram.vasudevan3/blog/2006/12/23/change-management-experience-from-a-sap-xi-architectural-study
Also have a look here,
http://help.sap.com/saphelp_erp2005vp/helpdata/en/93/a3a74046033913e10000000a155106/content.htm
http://help.sap.com/saphelp_erp2005vp/helpdata/en/94/c461bd75be534c89b5646b6ddc0aff/frameset.htm
Regards,
Prateek

Webcenter Sites Installation fail during deployment

Hi
I have been trying to do a proof of concerpt with Webcenter Site on Weblogic 10.3.6 on windows 7, it installs fine but failes with the following error during deployment
An error occurred during activation of changes, please see the log for details.
User-specified log class 'org.apache.commons.logging.impl.Log4JLogger' cannot be found or is not useable.:org.apache.commons.logging.LogConfigurationException:User-specified log class 'org.apache.commons.logging.impl.Log4JLogger' cannot be found or is not useable.
has anyone encountered this and how have they resolved this???
Thanks in advance
Thabo

I have managed to fix this problem, by adding the following in the weblogic.xml
<container-descriptor>
<prefer-web-inf-classes>true</prefer-web-inf-classes>
</container-descriptor>

Non production usage of the MCOS deployment

Hi everyone,
As we all know, it's time for HANA SPS09.
So recently I have read a lot about MDC(multitenant database containers). And I learn that by comparing to other architectures.(include SCOS, MCOD, MCOS).
However, when referring to MCOS, multiple component one system, it says no support for production.
So why can't MCOS use for production?
I know Cons of MCOS are performance issue and additional sizing. So it's just due to this?
If so, why the performance of MCOS is bad?
Looking forward to your kind help.
Thanks in advance.
Geo

Hello Geoffrey,
As you know, for MCOS, contention for resources may negatively impact performance (you can't control CPU usage between different databases).
For MDC, the CPU and memory resouce can be controlled by MDC.
In my view, SAP considers to reach the biggest performance for HANA and no resource contention between databases for production HANA system with available TCO.
Regards,
Ning Tong

Production site RAC redundant architecture deployment

Similar Messages

Maybe you are looking for